CN104702506B - A kind of message transmitting method, network node and message transfer system - Google Patents

Abstract

The embodiment of the present invention discloses a message transmission method, a network node and a message transmission system, wherein the centralized network node receives the message sent by the source edge routing bridge device through TRILL unicast encapsulation, wherein the above message is broadcast message or unknown unicast message or multicast message; the centralized network node determines the target edge routing bridge device set according to the multicast forwarding table, wherein the target routing bridge device set does not include the source edge routing bridge device, and the multicast The forwarding table includes the identification of each edge routing bridge device in the network where the above-mentioned centralized network node is located; the above-mentioned centralized network node sends the above-mentioned messages one by one to the routing bridges contained in the above-mentioned target edge routing bridge device set through TRILL unicast encapsulation equipment. The technical scheme provided by the invention can effectively save multicast resources and improve the expansion performance of the system.

Description

A message transmission method, network node and message transmission system

technical field

The invention relates to the communication field, in particular to a message transmission method, a network node and a message transmission system.

Background technique

Transparent Interconnection of Lots of Links (English: Transparent Interconnection of Lots of Link, abbreviation: TRILL) protocol is a routing protocol based on link state calculations. Devices running the TRILL protocol are called routing bridge devices (English: Route Bridge, abbreviation: RB ), the network composed of RB nodes is called TRILL network. The RB node in the TRILL network is identified by the alias (nickname) as the identifier (Identity, abbreviation: ID) of the RB node. The nickname must be unique in the entire network, that is, the nickname of each RB node is different.

In the TRILL network, the reverse path forwarding (English: Reverse Path Forwarding, abbreviation: RPF) check is used to prevent loops of broadcast, unknown unicast and multicast packets (for ease of description, broadcast packets, unknown unicast packets Or multicast packets are collectively referred to as BUM packets). The mechanism of the RPF check is that when the TRILL network side receives a BUM packet, it checks whether the BUM packet enters from the specified ingress port. Publish the item and copy and forward the BUM message, otherwise, the BUM message will be discarded.

Usually, a server or customer edge (English: Customer Edge, abbreviation: CE) device is dual-homed or multi-homed to multiple edge RB nodes through a Link Aggregation Group (English: Link Aggregation Group, abbreviation: LAG). The message randomly selects an uplink to access the TRILL network through the hash (Hash) algorithm. In order to prevent the remote edge RB node from jumping when learning Media Access Control (English: Media Access Control, abbreviation: MAC), it is necessary to Multiple edge RB nodes with dual-homing access or multi-homing access are virtualized into a vRB node, and a virtual alias (vnickname) is used to identify the vRB node. When dual-homing or multi-homing access to multiple edge RB nodes, the RPF check will cause the BUM packets entering from some ingress ports of the vRB nodes to be discarded. For example, as shown in Figure 1, it is assumed that the server or CE accesses RB1, RB2, and RB3 through multi-homing. RB1, RB2, and RB3 are virtualized into a vRB node identified as "vnickname1", and the root node RBx of the multicast distribution tree is identified by "vnickname root1". , generate the RPF checklist shown in Table 1 for vnickname1 on the RBx node:

Table 1

The server or CE selects a link to send a BUM message to the RBx node. Table 1 shows that only the BUM message entering from the port port1 of the RB1 node will be considered as a legal BUM message when the RPF check is performed at the RBx node. However, it is forwarded to the RBn node, and the BUM message entering from the port port2 of the RB2 node and the port port3 of the RB3 node will be discarded at the RBx because it cannot pass the RPF check.

In view of the above problems, a coordinated multicast tree (English: Coordinated MulticastTrees, abbreviation: CMT) scheme is proposed. The CMT scheme generates multiple multicast distribution trees for the actual physical nodes of the vRB. Different multicast distribution trees generate different multicast distribution trees. The RPF checklist is used to solve the problem of RPF check failure in dual-homing or multi-homing access. Based on the CMT scheme, as shown in Figure 2, three multicast distribution trees are generated for the RB1 node, RB2 node, and RB3 node, and the RPF checklist shown in Table 2 is generated for vnickname1 at the RBx node:

Table 2

In this way, when the RBx performs the RPF check, the problem of discarding the BUM packets entering from some ingress ports of the vRB node can be avoided.

However, it can be seen from the above CMT scheme that the number of multicast distribution trees must be at least equal to the actual number of nodes of the largest vRB node in the network. For example, four multicast distribution trees are required for four-homing access. When the number of multicast distribution trees increases, a large amount of multicast resources will also be consumed. For example, in a virtual local area network (English: Virtual Local Area Network, abbreviation: VLAN), the multicast resources that need to be provided by the device are: the number of VLANs *The number of multicast distribution trees; in the Internet Group Management Protocol (English: Internet Group Management Protocol, abbreviation: IGMP) snooping, that is, in the IGMP Snooping multicast constraint mechanism, the multicast resources that need to be provided by the device are the number of VLANs*group The number of broadcast distribution trees * the number of multicast groups. Since the multicast resources of the device are usually limited, if the multicast resources are consumed in large quantities due to the increase in the number of multicast distributions, the service provision capabilities (such as multicast services, number of VLANs, etc.) will also be restricted, resulting in The scalability of the system will also be constrained.

Contents of the invention

The invention provides a message transmission method, a network node and a message transmission system, which are used for saving multicast resources and improving the expansion performance of the system.

One aspect of the present invention provides a message transmission method, including:

The centralized network node receives the message sent by the source edge routing bridge device through the multi-link transparent interconnection TRILL unicast encapsulation mode, wherein the above-mentioned message is a broadcast message or an unknown unicast message or a multicast message;

The centralized network node determines the target edge routing bridge device set according to the multicast forwarding table, wherein the target routing bridge device set does not include the source edge routing bridge device, and the multicast forwarding table includes each The identification of the edge routing bridge device;

The centralized network node sends the above packets one by one to the routing bridge devices included in the target edge routing bridge device set through TRILL unicast encapsulation.

Based on the first aspect of the present invention, in a first possible implementation manner,

Before determining the target edge routing bridge device set according to the multicast forwarding table, the above method also includes:

Receive the link status message of each edge routing bridge device in the network where the centralized network node is located;

Generate the above multicast forwarding table according to the received link state message of each edge routing bridge device in the network where the above centralized network node is located.

Based on the first possible implementation of the first aspect of the present invention, in the second possible implementation, the above-mentioned link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device, the above-mentioned multicast The forwarding table specifically includes: the corresponding relationship between the identification of each edge routing bridge device in the network where the above-mentioned centralized network node is located and the virtual local area network identification;

The above-mentioned determination of the target edge routing bridge device set according to the above-mentioned multicast forwarding table includes:

Obtain the virtual local area network identification in the above-mentioned message, and determine the edge routing bridge device that accesses the virtual local area network where the above-mentioned message is located according to the above-mentioned virtual local area network identification and the above-mentioned multicast forwarding table;

Wherein, the target edge routing bridge device set is composed of the determined edge routing bridge devices that access the virtual local area network where the packet is located.

Based on the first possible implementation of the first aspect of the present invention, in the third possible implementation, the above-mentioned link state message includes: the VLAN ID of the virtual local area network accessed by the edge routing bridge device and the edge routing bridge The identifier of the layer 2 multicast domain that the device joins, the above multicast forwarding table specifically includes: the identifier of each edge routing bridge device in the network where the above centralized network node is located, the virtual local area network identifier, and the layer 2 group that the edge routing bridge device joins The corresponding relationship between broadcast domain identifiers;

The above-mentioned determination of the target edge routing bridge device set according to the multicast forwarding table includes:

Obtain the virtual local area network identifier and the identifier of the layer 2 multicast domain in the above message, use the combination of the above virtual local area network identifier and the identifier of the layer 2 multicast domain as an index, and determine the location where the above message is accessed according to the above multicast forwarding table and join the edge routing bridge device of the Layer 2 multicast domain accessed by the above message;

Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the message is located and join the Layer 2 multicast domain that the message accesses.

Based on the first aspect of the present invention, or the first possible implementation of the first aspect of the present invention, or the second possible implementation of the first aspect of the present invention, or, the third possible implementation of the first aspect of the present invention Implementation manner, in a fourth possible implementation manner, the above method further includes:

The above-mentioned centralized network node floods the extended link state message in the above-mentioned network, and the above-mentioned extended link state message carries the attribute and the super-level name of the above-mentioned centralized network node. The above-mentioned attribute indicates that the above-mentioned centralized network node supports the centralized replication function, and the above-mentioned super-level name It is the alias used by the above centralized network nodes in the centralized replication scenario.

Based on the fourth possible implementation of the first aspect of the present invention, in the fifth possible implementation, the centralized network node receives the message sent by the source edge routing bridge device through the TRILL unicast encapsulation mode of transparent interconnection of multiple links , followed by:

Judging whether the egress alias of the message sent by the above-mentioned TRILL unicast encapsulation method is the above-mentioned super-level name, if the above-mentioned egress alias is the above-mentioned super-level name, perform the above-mentioned steps of determining the set of target edge routing bridge devices according to the multicast forwarding table.

Based on the first aspect of the present invention, or the first possible implementation of the first aspect of the present invention, or the second possible implementation of the first aspect of the present invention, or, the third possible implementation of the first aspect of the present invention Implementation manner, in the sixth possible implementation manner, if the target edge routing bridge device set includes a virtual routing bridge device virtualized by multiple routing bridge devices with dual-homing access or multi-homing access, then the above-mentioned will The above messages are sent one by one to the routing bridge devices included in the target edge routing bridge device set, including:

Send the above-mentioned message to one of the routing bridge devices of the above-mentioned dual-homing access or multi-homing access; or

The above message is sent to the virtual routing bridge device composed of multiple routing bridge devices with dual-homing access or multi-homing access.

A second aspect of the present invention provides a network node, including:

The first receiving unit is used to receive the message sent by the source edge routing bridge device through the multi-link transparent interconnection TRILL unicast encapsulation mode, wherein the above message is a broadcast message or an unknown unicast message or a multicast message;

A determination unit, configured to determine a target edge routing bridge device set according to a multicast forwarding table, wherein the target routing bridge device set does not include the source edge routing bridge device, and the multicast forwarding table includes each The identification of the edge routing bridge device;

The first sending unit is configured to send the packets received by the receiving unit one by one to the routing bridge devices included in the target edge routing bridge device set determined by the determining unit through TRILL unicast encapsulation.

Based on the second aspect of the present invention, in a first possible implementation manner,

The above network nodes also include:

The second receiving unit is configured to receive the link state message of each edge routing bridge device in the network where the centralized network node is located;

The generating unit is configured to generate the above-mentioned multicast forwarding table according to the link state message of each edge routing bridge device in the network where the above-mentioned network node is located and received by the above-mentioned second receiving unit.

Based on the first possible implementation of the second aspect of the present invention, in the second possible implementation, the link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device, the above multicast The forwarding table specifically includes: the corresponding relationship between the identification of each edge routing bridge device in the network where the above-mentioned centralized network node is located and the virtual local area network identification;

The determining unit is specifically used to: obtain the virtual local area network identifier in the above message, and determine the edge routing bridge device that accesses the virtual local area network where the above message is located according to the above virtual local area network identifier and the above multicast forwarding table;

Wherein, the target edge routing bridge device set is composed of the determined edge routing bridge devices that access the virtual local area network where the packet is located.

Based on the first possible implementation of the second aspect of the present invention, in the third possible implementation, the above-mentioned link state message includes: the VLAN ID of the virtual local area network accessed by the edge routing bridge device and the edge routing bridge The identifier of the Layer 2 multicast domain that the device accesses, the above-mentioned multicast forwarding table specifically includes: the identifier of each edge routing bridge device in the network where the above-mentioned centralized network node is located, the virtual local area network identifier, and the Layer 2 multicast domain that the edge routing bridge device joins. The corresponding relationship of the identification of the multicast domain;

The above-mentioned determination unit is specifically used to: obtain the virtual local area network identifier and the identifier of the layer-2 multicast domain in the above-mentioned message, and use the combination of the above-mentioned virtual local area network identifier and the identifier of the layer-2 multicast domain as an index, according to the above-mentioned multicast forwarding table Determine the edge routing bridge device that accesses the virtual local area network where the above message is located and joins the Layer 2 multicast domain that the above message accesses;

Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the message is located and join the Layer 2 multicast domain that the message accesses.

Based on the second aspect of the present invention, or the first possible implementation of the second aspect of the present invention, or the second possible implementation of the second aspect of the present invention, or the third possible implementation of the second aspect of the present invention , in a fourth possible implementation manner, the above network node further includes:

The second sending unit is used to flood the extended link state message in the above network, the above extended link state message carries the attribute and super class name of the above centralized network node, and the above attribute indicates that the above mentioned centralized network node supports the centralized replication function, The above-mentioned super-level name is the alias used by the above-mentioned centralized network nodes in the centralized replication scenario.

Based on the fourth possible implementation manner of the second aspect of the present invention, in a fifth possible implementation manner, the foregoing network node further includes:

The judging unit is used to judge whether the egress alias of the message sent by the TRILL unicast encapsulation method is the above superclass name;

The determining unit is triggered when the judging result of the judging unit is yes.

Based on the second aspect of the present invention, or the first possible implementation of the second aspect of the present invention, or the second possible implementation of the second aspect of the present invention, or the third possible implementation of the second aspect of the present invention , in a sixth possible implementation manner, the above-mentioned first sending unit is further configured to: when the above-mentioned target edge routing bridge device set includes a virtual network virtualized by multiple routing bridge devices with dual-homing access or multi-homing access In the case of a routing bridge device, send the above message to one of the routing bridge devices of the above-mentioned dual-homed or multi-homed access, or send the above-mentioned message to the above-mentioned dual-homed or multi-homed The virtual routing bridge device composed of two routing bridge devices sends the above message.

A third aspect of the present invention provides a message transmission system, including:

Concentrate network nodes and more than two edge routing bridge devices;

Wherein, the above-mentioned centralized network node is in the same network as the above-mentioned two or more edge routing bridge devices;

The aforementioned centralized network node is any one of the aforementioned network nodes in the second aspect of the present invention.

As can be seen from the above, the source edge routing bridge device in the present invention sends messages to the centralized network node through the TRILL unicast encapsulation mode, and the centralized network node sends the received messages to the target edge routing bridge one by one through the TRILL unicast encapsulation mode For the routing bridge devices included in the device set, since the target routing bridge device set does not include the above-mentioned source edge routing bridge devices, it is not necessary to pass the RPF check to ensure that there will be no loops during the transmission of the message, and the centralized network node single The method of broadcast receiving and unicast sending can effectively reduce the number of multicast distribution trees, thereby saving multicast resources. Under the condition of the same hardware multicast resources, the system can support more multicast services and improve the system efficiency. Extend performance.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.

Figure 1 is a schematic diagram of the application scenario structure of the existing TRILL network;

Figure 2 is a schematic structural diagram of a TRILL network application scenario based on the CMT scheme;

Fig. 3 is a schematic flow chart of an embodiment of the message transmission method provided by the present invention;

Figure 4-a is a schematic structural diagram of a TRILL network application scenario based on the message transmission method provided by the present invention;

Figure 4-b is a schematic diagram of a message transmission flow under the TRILL network shown in Figure 4-a based on the message transmission method provided by the present invention;

Figure 4-c is a schematic diagram of another message transmission flow under the TRILL network shown in Figure 4-a based on the message transmission method provided by the present invention;

Figure 4-d is a schematic diagram of another message transmission flow under the TRILL network shown in Figure 4-a based on the message transmission method provided by the present invention;

Figure 5-a is a schematic structural diagram of another TRILL network application scenario based on the message transmission method provided by the present invention;

Figure 5-b is a schematic diagram of a message transmission flow under the TRILL network shown in Figure 5-a based on the message transmission method provided by the present invention;

FIG. 6 is a schematic structural diagram of an embodiment of a network node provided by the present invention;

FIG. 7 is a schematic structural diagram of another embodiment of a network node provided by the present invention;

FIG. 8 is a schematic structural diagram of another embodiment of a network node provided by the present invention;

FIG. 9 is a schematic structural diagram of an embodiment of a message transmission system provided by the present invention.

Detailed ways

Embodiments of the present invention provide a message transmission method, a network node and a message transmission system.

In order to make the purpose, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described The embodiments are only some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, various other embodiments obtained by persons of ordinary skill in the art without making creative efforts all belong to the protection scope of the present invention.

A message transmission method provided by the embodiment of the present invention is described below, please refer to FIG. 3, the message transmission method in the embodiment of the present invention includes:

301. The centralized network node receives the message sent by the source edge RB through TRILL unicast encapsulation.

In the embodiment of the present invention, the edge RB refers to the RB set on the edge of the TRILL network. When the edge RB receives broadcast, unknown unicast and multicast messages (ie BUM messages) from the server or customer edge device (English: Customer Edge, abbreviation: CE), and knows that the above-mentioned centralized network nodes support the centralized replication function , the edge RB sends the received BUM message to the centralized network node through TRILL unicast encapsulation. Specifically, the edge RB encapsulates the unicast TRILL tunnel header on the BUM message. The unicast TRILL tunnel header includes the source nickname and the target nickname, where the source nickname is the nickname of the edge RB, and the target nickname is the super nickname of the above-mentioned centralized network node.

It should be noted that the above centralized network nodes have a common nickname and a super nickname in the network. The normal nickname above works the same as the nickname for other devices. The above super nickname is an alias used by the centralized network nodes in the centralized replication scenario, which is specially used for the centralized replication function. The above-mentioned centralized network node floods the extended link state message in the above-mentioned network, and the extended link-state message flooded by the above-mentioned centralized network node carries the attribute of the above-mentioned centralized network node and the above-mentioned super nickname, and the above-mentioned attribute indicates that the above-mentioned centralized network node Support the centralized replication function, so that when the edge RB listens to the extended link status message flooded by the centralized network node, it can know that the centralized network node supports the centralized replication function and the alias used by the centralized network node in the centralized replication scenario.

The above centralized replication scheme refers to: the edge RB node sends a BUM message to the centralized network node through the TRILL unicast encapsulation mode instead of the multicast mode, and when the above-mentioned centralized network node receives the BUM message, the The target nickname in the unicast TRILL tunnel header can determine that the above message is a message that requires centralized replication, and the above centralized network node sends the BUM message to other edge RBs through multiple unicast messages in a centralized replication manner.

Optionally, in the embodiment of the present invention, the link state message sent by the edge RB is extended, and the extended link state message sent by the edge RB carries the centralized replication function indicator of the edge RB. The centralized network node determines whether the edge RB has the centralized replication function by monitoring the centralized replication function indicator carried in the extended link state message sent by the edge RB.

Among them, it should be noted that if the server or CE accesses the edge RB through dual-homed access or multi-homed access, the source nickname in the above-mentioned unicast TRILL tunnel header should be: ) The vnickname of the virtual RB virtualized by multiple edge RBs with dual-homing access or multi-homing access, and the target nickname is still the super nickname of the above centralized network node.

302. The centralized network node determines a target edge RB set according to the multicast forwarding table.

Wherein, the target RB set does not include the source edge RB, and the multicast forwarding table includes the identifier of each edge RB in the network where the centralized network node is located.

The multicast forwarding table in the embodiment of the present invention can be pre-received by the centralized network node above the link status message of each edge RB in the network where the centralized network node is located, and according to the received link status message of each edge RB Status message generation. Therefore, the multicast forwarding table must include the source edge RBs. In this case, the centralized network node prunes the source edge RBs so that the determined set of target edge RBs does not include the source edge RBs.

In an application scenario, the centralized network node monitors the link state message of each edge RB in the network where it is located, wherein the link state message contains the VLAN ID of the VLAN accessed by the edge RB, and the centralized network node The node generates a multicast forwarding table according to the monitored link state message of each edge RB, and the multicast forwarding table includes: the corresponding relationship between the identifier of each edge RB in the network where the centralized network node is located and the VLAN identifier. When the above-mentioned centralized network node receives the message sent by the source edge RB through TRILL unicast encapsulation, it obtains the VLAN ID in the message, and determines the VLAN that accesses the above-mentioned message according to the above-mentioned VLAN ID and the above-mentioned multicast forwarding table. Edge RBs, the target edge RB set is composed of the determined edge RBs accessing the VLAN where the packet is located.

In another application scenario, the above-mentioned centralized network node monitors the link state message of each edge RB in the network where it is located, wherein the link state message includes the VLAN ID of the VLAN accessed by the edge RB and the edge RB The identifier of the layer-2 multicast domain to join, the above-mentioned centralized network node generates a multicast forwarding table according to the monitored link state message of each of the above-mentioned edge RBs, and the multicast forwarding table includes: The corresponding relationship between the ID of each edge RB, the ID of the VLAN, and the ID of the Layer 2 multicast domain to which the edge RB joins. When the above-mentioned centralized network node receives the message sent by the source edge RB through TRILL unicast encapsulation, it obtains the VLAN identifier and the identifier of the two-layer multicast domain in the packet, and uses the combination of the VLAN identifier and the identifier of the two-layer multicast domain Combined into an index, determine the VLAN where the message is accessed according to the above multicast forwarding table, and add the edge RB of the Layer 2 multicast domain to which the message is accessed, and the set of target edge RBs is determined by the above-mentioned access to the message The VLAN where the message is located, and added to the edge RB of the Layer 2 multicast domain that the message accesses.

It should be noted that the multicast forwarding table in the embodiment of the present invention may also be generated by other devices or in other ways and sent to the centralized network node in a wireless or wired manner, which is not limited here.

Optionally, after the centralized network node receives the message sent by the source edge RB through TRILL unicast encapsulation in step 301, it is judged whether the egress alias of the message sent by the above-mentioned TRILL unicast encapsulation is the above-mentioned super-level name, if , execute step 302 and subsequent steps, if not, end this process, or discard the message or process the message according to the traditional message processing process (such as the CMT scheme), which is not limited here.

303. The centralized network node sends the above-mentioned packets one by one to the RBs included in the above-mentioned target edge RB set through TRILL unicast encapsulation.

After the target edge RB set is determined in step 302, the above-mentioned centralized network node sends the above-mentioned packets one by one to the RBs included in the above-mentioned target edge RB set through TRILL unicast encapsulation.

Further, if the set of target edge RBs includes virtual RBs virtualized by multiple RBs with dual-homed access or multi-homed access, the centralized network node sends RBs to the above-mentioned dual-homed access or multi-homed RBs through TRILL unicast encapsulation. One of the multiple RBs that are homed to access sends the above-mentioned message, or sends the above-mentioned message to a virtual RB composed of multiple RBs that are dual-homed or multi-homed.

Optionally, during the process that the above-mentioned centralized network node sends the above-mentioned message to one of the multiple RBs with dual-homed access or multi-homed access through TRILL unicast encapsulation, the above-mentioned centralized network node uses Equal-cost multipath (ECMP, Equal-Cost Multipath Routing) mechanism, select a link from the multiple links from the above-mentioned centralized network node to the multiple edge RBs contained in the above-mentioned virtual RB, and send the above-mentioned message to the selected link The edge RB on the road, or the above-mentioned centralized network node may also select an edge RB from multiple edge RBs included in the above-mentioned virtual RB through other mechanisms or randomly to send the message, which is not limited here.

Optionally, when the centralized network node sends the above-mentioned message to the RBs included in the target edge RB set, keep the source nickname value in the above-mentioned message unchanged, that is, keep the source nickname value as the nickname of the above-mentioned source edge RB, so that When the RBs in the target edge RB set receive the message, the learned MAC entry is the nickname of the source edge RB, not the nickname of the centralized network node.

It should be noted that the centralized network node in the embodiment of the present invention is an RB node.

It can be seen from the above that in the embodiment of the present invention, the source edge RB sends messages to the centralized network node through the TRILL unicast encapsulation method, and the centralized network node sends the received messages to the target edge RB set one by one through the TRILL unicast encapsulation method For the included RBs, since the target RB set does not include the above-mentioned source edge RBs, there is no need to pass the RPF check to ensure that there will be no loops during the transmission of the message, and the centralized network node unicast reception and unicast transmission method It can effectively reduce the number of multicast distribution trees, thereby saving multicast resources. Under the condition of the same hardware multicast resources, the system can support more multicast services and improve the expansion performance of the system.

The message transmission method in the embodiment of the present invention is described below in a specific application scenario. Assume that there is a TRILL network topology as shown in Figure 4-a. In Figure 4-a, RB1~RB6 are edge RBs, RBx is a centralized network node, and RBx and RB1~RB6 are connected to the same network. Suppose RB1~RB6 The aliases of RBx are nickname1~nickname6, and the superlevel name of RBx is nicknameroot. Server 1 is dual-homed to RB1 and RB2, and RB1 and RB2 are virtualized into a logical vRB, assuming that the alias of vRB is vnickname. In the embodiment of the present invention, RBx, RB1-RB6 all support the centralized replication function.

RBx monitors the link state messages of RB1-RB6, and generates a multicast forwarding table as shown in Table 3 according to the link state messages:

table 3

In the embodiment of the present invention, server 1 randomly selects an uplink to send a BUM message through a HASH algorithm, assuming that RB1 is selected to send the BUM message. RB1 learns that RBx supports the centralized replication function by listening to the centralized network node attributes carried in the extended link state message flooded by RBx. After RB1 encapsulates the unicast TRILL tunnel header in the BUM message, it encapsulates the The BUM message is sent to RBx, and the format of the unicast TRILL tunnel header is shown in Table 4:

Table 4

source nickname target nickname vnickname nickname root

RBx receives the BUM message from RB1, and according to the multicast forwarding table shown in Table 3, determines that the set of target edge RBs is RB3~RB6 (vRB is the source RB of the BUM message, so the pruning process makes the set of target edge RBs excluding vRB), RBx sends the BUM message to RB3-RB6 one by one through TRILL unicast encapsulation.

The transmission flow of the BUM message is shown by the arrow in Figure 4-b.

The message transmission method in the embodiment of the present invention is described below in another specific application scenario. Assume that there is a TRILL network topology as shown in Figure 4-a. In Figure 4-a, RB1~RB6 are edge RBs, RBx is a centralized network node, and RBx and RB1~RB6 are connected to the same network. Suppose RB1~RB6 The aliases of RBx are nickname1~nickname6, and the superlevel name of RBx is nicknameroot. Server 1 is dual-homed to RB1 and RB2, and RB1 and RB2 are virtualized into a logical vRB, assuming that the alias of vRB is vnickname. In the embodiment of the present invention, RBx, RB1-RB6 all support the centralized replication function, assuming that there are currently four VLANs, which are identified by VLAN1-VLAN4 respectively, RB1 and RB2 are connected to VLAN1, RB3 is connected to VLAN2, and RB4 is connected to VLAN3. RB5 is connected to VLAN1, VLAN2, and VLAN3, and RB6 is connected to VLAN1, VLAN2, VLAN3, and VLAN4. The link status messages sent by RB1-RB6 include the VLAN IDs of the VLANs they are connected to.

RBx monitors the link status messages of RB1-RB6, and generates a multicast forwarding table as shown in Table 5 according to the link status messages:

table 5

In the embodiment of the present invention, server 1 randomly selects an uplink to send a BUM message through a HASH algorithm. Assuming that RB1 is selected to send the BUM message, the ID of the VLAN network where the BUM message is located is VLAN1. Extend the attribute of the centralized network node carried in the link state message to learn that the RBx supports the centralized replication function. After RB1 encapsulates the unicast TRILL tunnel header on the BUM message, it sends the BUM message to RBx through TRILL unicast encapsulation. The format of the unicast TRILL tunnel header is shown in Table 4.

RBx receives the BUM message from RB1, and determines the target edge RB set as RB5 and RB6 according to the multicast forwarding table shown in Table 5 (vRB is the source RB of the BUM message, so the pruning process makes the target edge RB set excluding vRB), RBx sends the BUM message to RB5 and RB6 one by one through TRILL unicast encapsulation.

The transmission flow of the BUM message is shown by the arrow in Figure 4-c.

The message transmission method in the embodiment of the present invention is described below in another specific application scenario. Assume that there is a TRILL network topology as shown in Figure 4-a. In Figure 4-a, RB1~RB6 are edge RBs, RBx is a centralized network node, and RBx and RB1~RB6 are connected to the same network. Suppose RB1~RB6 The aliases of RBx are nickname1~nickname6, and the superlevel name of RBx is nicknameroot. Server 1 is dual-homed to RB1 and RB2, and RB1 and RB2 are virtualized into a logical vRB, assuming that the alias of vRB is vnickname. In this embodiment of the present invention, RBx, RB1-RB6 all support the centralized replication function. Assume that there are currently four VLANs, which are identified by VLAN1-VLAN4 respectively, and each VLAN contains two Layer 2 multicast domains. The identifiers of the two Layer 2 multicast domains are GMAC1 and GMAC2 respectively, the identifiers of the two Layer 2 multicast domains of VLAN2 are GMAC3 and GMAC4 respectively, the identifiers of the two Layer 2 multicast domains of VLAN3 are GMAC5 and GMAC6 respectively, and the identifiers of the two Layer 2 multicast domains of The identifiers of the two Layer 2 multicast domains are GMAC7 and GMAC8 respectively. RB1 and RB2 join GMAC1 and GMAC2 of VLAN1; RB3 joins GMAC3 of VLAN2; RB4 joins GMAC5 and GMAC6 of VLAN3; RB5 joins GMAC1 of VLAN1, GMAC4 of VLAN2 and GMAC5 of VLAN3; GMAC6, GMAC7 and GMAC8 of VLAN4. The link state messages sent by RB1-RB6 include the VLAN identifiers of the VLANs they access and the identifiers of the Layer 2 multicast domains they join.

RBx monitors the link status messages of RB1~RB6, and generates the multicast forwarding table shown in Table 6 according to the link status messages:

Table 6

In the embodiment of the present invention, server 1 randomly selects an uplink to send a BUM message through a HASH algorithm. Suppose that RB1 is selected to send the BUM message. The BUM message is located in GMAC2 of VLAN1. According to the tree structure of the centralized network nodes carried in the road status message, the RBx supports the centralized replication function. After RB1 encapsulates the unicast TRILL tunnel header on the BUM message, it sends the BUM message to RBx through TRILL unicast encapsulation. The format of broadcast TRILL tunnel header is shown in Table 4.

RBx receives the BUM message from RB1, and determines that the target edge RB set is RB6 according to the multicast forwarding table shown in Table 6 (vRB is the source RB of the BUM message, so the pruning process makes the target edge RB set not include vRB), RBx sends the BUM message to RB6 through TRILL unicast encapsulation.

The transmission flow of the BUM message is shown by the arrow in Figure 4-d.

The message transmission method in the embodiment of the present invention is described below in another specific application scenario. Assume that there is a TRILL network topology as shown in Figure 5-a. In Figure 5-a, RB1~RB6 are edge RBs, RBx is a centralized network node, and RBx and RB1~RB6 are connected to the same network. Suppose RB1~RB6 The aliases of RBx are nickname1~nickname6, and the superlevel name of RBx is nicknameroot. Server 1 is dual-homed to RB1 and RB2, and RB1 and RB2 are virtualized into a logical vRB. Assume that the alias of vRB is vnickname, and server 2 is connected to RB6. In the embodiment of the present invention, RBx, RB1-RB6 all support the centralized replication function. RBx monitors the link status messages of RB1-RB6, and generates a multicast forwarding table as shown in Table 3 according to the link status messages.

In the embodiment of the present invention, the server 2 sends the BUM message through RB6. RB6 learns that the RBx supports the centralized replication function by listening to the centralized network node attributes carried in the extended link state message flooded by the RBx. After RB6 encapsulates the unicast TRILL tunnel header in the BUM message, it encapsulates the The BUM message is sent to RBx, and the format of the unicast TRILL tunnel header is shown in Table 7:

Table 7

source nickname target nickname nickname6 nickname root

RBx receives the BUM message from RB6, and determines the set of target edge RBs as vRB, RB3, RB4, and RB5 according to the multicast forwarding table shown in Table 3 (RB6 is the source RB of the BUM message, because of the pruning process, making The target edge RB set does not include RB6), RBx sends the BUM message to RB3, RB4 and RB5 one by one through unicast, since vRB contains two links, pointing to RB1 and RB2 respectively, at this time, RBx uses the ECMP mechanism or Randomly select one of the links to send a BUM message. Assuming that the link pointing to RB1 is selected, RBx sends the BUM message to RB1 through TRILL unicast encapsulation, and RB1 decapsulates the TRILL header of the received BUM message Transfer processing to server 1 is performed.

The transmission flow of the BUM message is shown by the arrow in Figure 5-b.

The embodiment of the present invention also provides a network node, please refer to FIG. 6, the network node 600 in the embodiment of the present invention includes:

The first receiving unit 601 is configured to receive a message sent by the source edge routing bridge device through TRILL unicast encapsulation, wherein the above message is a broadcast message or an unknown unicast message or a multicast message;

The determining unit 602 is configured to determine a target edge routing bridge device set according to the multicast forwarding table, wherein the target routing bridge device set does not include the source edge routing bridge device, and the multicast forwarding table includes each An ID of an edge routing bridge device;

The first sending unit 603 is configured to send the packets received by the first receiving unit 601 one by one to the RBridge devices included in the target edge RBridge device set determined by the determining unit 602 through TRILL unicast encapsulation.

On the basis of the embodiment shown in FIG. 6, as shown in FIG. 7, the network node 700 further includes:

The second receiving unit 604 is configured to receive a link state message of each edge routing bridge device in the network where the network node 700 is located;

The generating unit 605 is configured to generate a multicast forwarding table according to the link state message of each edge routing bridge device in the network where the network node 700 is located received by the second receiving unit 604 .

In an application scenario, the above link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device, and the multicast forwarding table generated by the generating unit 605 also includes: The corresponding relationship between the identification of each edge routing bridge device and the virtual local area network identification; the determination unit 602 is specifically used to: obtain the virtual local area network identification in the above-mentioned message, and determine the access according to the above-mentioned virtual local area network identification and the multicast forwarding table generated by the generation unit 605 The edge routing bridge device of the virtual local area network where the above packet is located; wherein, the target edge routing bridge device set is formed by the edge routing bridge device determined by the determining unit 602 .

In another application scenario, the above link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device and the identifier of the layer 2 multicast domain accessed by the edge routing bridge device, then the generation unit 605 generates The multicast forwarding table also includes: the corresponding relationship between the identifier of each edge routing bridge device in the network where the network node 600 is located, the virtual local area network identifier, and the identifier of the layer 2 multicast domain joined by the edge routing bridge device; the determining unit 602 specifically uses In: obtaining the virtual local area network identification and the identification of the layer-2 multicast domain in the above-mentioned message, using the combination of the identification of the above-mentioned virtual local area network and the identification of the layer-2 multicast domain as an index, and determining according to the multicast forwarding table generated by the generation unit 605 An edge routing bridge device that accesses the virtual local area network where the above-mentioned message is located and joins the Layer 2 multicast domain to which the above-mentioned message is accessed;

Further, if the target edge routing bridge device set includes a virtual routing bridge device virtualized by multiple routing bridge devices with dual-homed access or multi-homed access, the first sending unit 603 is also used to encapsulate through TRILL unicast way, to send the above-mentioned message to one of the routing bridge devices of the above-mentioned dual-homed or multi-homed access, or to send the above-mentioned dual-homed or multi-homed The virtual routing bridge device composed of devices sends the above message.

Optionally, during the process in which the above network node sends the above message to one of the multiple routing bridge devices with dual-homing access or multi-homing access through the first sending unit 603, the above network node Using the ECMP mechanism, select a link from the multiple links of the above-mentioned network node to the multiple edge routing bridge devices included in the above-mentioned virtual routing bridge device, and send the above-mentioned message to the selected link through the first sending unit 603 Alternatively, the above-mentioned network node may select an edge routing bridge device from multiple edge routing bridge devices included in the virtual routing bridge device through other mechanisms or randomly, and send the selected edge routing bridge device to the selected edge routing bridge device through the first sending unit 603. The router device is used to send the message, which is not limited here.

Optionally, when the first sending unit 603 sends the above message to the RBs included in the target edge RB set, keep the source nickname value in the above message unchanged, that is, keep the source nickname value as the nickname of the source edge RB , so that when the RBs in the target edge RB set receive the packet, the learned MAC entry is the nickname of the source edge RB, rather than the nickname of the network node.

It should be noted that the network node in the embodiment of the present invention has a common nickname and a super nickname in the network. The normal nickname above works the same as the nickname for other devices. The above super nickname is the alias used by the above network nodes in the centralized replication scenario, which is specially used for the centralized replication function. Optionally, on the basis of the embodiment shown in FIG. 6 or FIG. 7 , the network node further includes: a second sending unit, configured to flood the extended link state message in the above network, wherein the above extended link state The message carries the attribute of the network node and a super-level name, the attribute indicates that the network node supports the centralized replication function, and the super-level name is an alias used by the network node in the centralized replication scenario. Further, the above-mentioned network node further includes: a judging unit, configured to judge whether the egress alias of the packet sent by the TRILL unicast encapsulation method is is the above-mentioned super-level name, the determining unit 602 is triggered when the judgment result of the above-mentioned judging unit is yes.

It should be noted that the network nodes in the embodiments of the present invention can be like the centralized network nodes in the above method embodiments, and can be used to implement all the technical solutions in the above method embodiments, and the functions of each functional module can be implemented according to the above method The method in the example is specifically implemented, and the specific implementation process may refer to the relevant description in the above-mentioned embodiments, and details are not repeated here.

It can be seen from the above that in the embodiment of the present invention, the source edge RB sends messages to the centralized network node through the TRILL unicast encapsulation method, and the centralized network node sends the received messages to the target edge RB set one by one through the TRILL unicast encapsulation method For the included RBs, since the target RB set does not include the above-mentioned source edge RBs, there is no need to pass the RPF check to ensure that there will be no loops during the transmission of the message, and the centralized network node unicast reception and unicast transmission method It can effectively reduce the number of multicast distribution trees, thereby saving multicast resources. Under the condition of the same hardware multicast resources, the system can support more multicast services and improve the expansion performance of the system.

Another network node in the implementation of the present invention is described below, please refer to FIG. 8. The network node 800 in the embodiment of the present invention includes:

An input device 801, an output device 802, a memory 803, and a processor 804 (the number of processors 804 in the network node 800 may be one or more, one processor in FIG. 8 is taken as an example). In some embodiments of the present invention, the input device 801 , the output device 802 , the memory 803 and the processor 804 may be connected through a bus or in other ways, as shown in FIG. 8 , the connection through a bus is taken as an example. Wherein, the memory 803 is used to store the data input from the input device 801, and may also store information such as necessary files for the processor 804 to process the data; the input device 801 and the output device 802 may include a port for the user device 800 to communicate with other devices, And it may also include output devices external to the user equipment 800 such as a display, keyboard, mouse, and printer. Specifically, the input device 801 may include a mouse and a keyboard, and the output device 802 may include a display. In this embodiment, the input device 801 The port for communicating with other devices in the output device 802 may be an antenna and/or a cable.

Wherein, the processor 804 performs the following steps:

Receive the message sent by the source edge RB through TRILL unicast encapsulation, wherein the above message is a BUM message;

Determine the target edge RB set according to the multicast forwarding table, wherein the target RB set does not include the source edge RB, and the multicast forwarding table includes the identifier of each edge RB in the network where the network node is located;

Through TRILL unicast encapsulation, the above packets are sent one by one to the RBs included in the above target edge RB set.

Optionally, the processor 804 is further configured to: receive a link state message of each edge RB in the network where the network node 800 is located, and generate a multicast forwarding table according to the received link state message of each edge RB .

In one application scenario, the above link state message includes: the VLAN ID of the VLAN accessed by the edge RB, and the multicast forwarding table generated by the processor 804 also includes: the ID of each edge RB in the network where the network node 800 is located The corresponding relationship between the identifier and the VLAN identifier; the processor 804 is specifically configured to: when receiving a packet sent by the source edge RB through TRILL unicast encapsulation, obtain the VLAN identifier in the packet, and obtain the VLAN identifier in the packet according to the above-mentioned VLAN identifier and the above-mentioned multicast transfer Publish the edge RB that is determined to access the VLAN where the above packet is located. Wherein, the target edge RB set is composed of the determined edge RBs accessing the VLAN where the packet is located.

In another application scenario, the above-mentioned link state message includes the VLAN identifier of the VLAN accessed by the edge RB and the identifier of the Layer 2 multicast domain joined by the edge RB, and the multicast forwarding table generated by the processor 804 also includes : The corresponding relationship between the identifier of each edge RB in the network where the network node 800 is located, the VLAN identifier, and the identifier of the layer 2 multicast domain joined by the edge RB; the processor 804 is specifically used for: when the network node 800 receives the source edge RB through TRILL When sending a packet in unicast encapsulation mode, obtain the VLAN identifier and the identifier of the layer 2 multicast domain in the packet, use the combination of the VLAN identifier and the identifier of the layer 2 multicast domain as an index, and determine according to the above multicast forwarding table Access the VLAN where the packet is located, and join the edge RB of the Layer 2 multicast domain to which the packet is accessed. Wherein, the target edge RB set is composed of the determined edge RBs that access the VLAN where the message resides and join the Layer 2 multicast domain that the message accesses.

Further, if the target edge RB set includes virtual RBs virtualized by multiple RBs of dual-homed access or multi-homed access, the processor 804 is further configured to: use TRILL unicast encapsulation to send RBs to the above-mentioned dual-homed access Either one of the multiple RBs with multi-homed access sends the above-mentioned message, or, the above-mentioned message is sent to a virtual RB composed of multiple RBs with dual-homed access or multi-homed access.

Optionally, when the processor 804 sends the above-mentioned message to one of the multiple RBs of the above-mentioned dual-homed access or multi-homed access through TRILL unicast encapsulation, the processor 804 uses ECMP mechanism, select a link from the multiple links from the network node 800 to the multiple edge RBs included in the above-mentioned virtual RB, and send the above-mentioned message to the edge RB on the selected link, or, the processor 804 can also adopt An edge RB is selected randomly from the plurality of edge RBs included in the virtual RB to send the message through other mechanisms, which is not limited here.

Optionally, when the processor 804 sends the above message to the RBs included in the target edge RB set, keep the source nickname value in the above message unchanged, that is, keep the source nickname value as the nickname of the source edge RB, so that When the RBs in the target edge RB set receive the message, the MAC entry learned is the nickname of the source edge RB, not the nickname of the network node 800 .

It should be noted that the network node in the embodiment of the present invention has a common nickname and a super nickname in the network. The normal nickname above works the same as the nickname for other devices. The above super nickname is the alias used by the above network nodes in the centralized replication scenario, which is specially used for the centralized replication function. Optionally, the processor 804 is further configured to: flood the extended link state message in the above network, where the above extended link state message carries the attribute and super class name of the above network node, and the above attribute indicates that the above network node supports Centralized replication function, the above-mentioned super-level name is an alias used by the above-mentioned network nodes in the centralized replication scenario. Further, the processor 804 is also used to: when receiving a message sent by the source edge RB through TRILL unicast encapsulation, determine the above-mentioned TRILL unicast Whether the egress alias of the message sent in the broadcast encapsulation mode is the above-mentioned super-level name, if the above-mentioned egress alias is the above-mentioned super-level name, the above-mentioned action and subsequent actions of determining the target edge RB set according to the multicast forwarding table are performed.

It should be noted that the network nodes in the embodiments of the present invention can be like the centralized network nodes in the above method embodiments, and can be used to implement all the technical solutions in the above method embodiments, and the functions of each functional module can be implemented according to the above method The method in the example is specifically implemented, and the specific implementation process may refer to the relevant description in the above-mentioned embodiments, and details are not repeated here.

It can be seen from the above that in the embodiment of the present invention, the source edge RB sends messages to the centralized network node through the TRILL unicast encapsulation method, and the centralized network node sends the received messages to the target edge RB set one by one through the TRILL unicast encapsulation method For the included RBs, since the target RB set does not include the above-mentioned source edge RBs, there is no need to pass the RPF check to ensure that there will be no loops during the transmission of the message, and the centralized network node unicast reception and unicast transmission method It can effectively reduce the number of multicast distribution trees, thereby saving multicast resources. Under the condition of the same hardware multicast resources, the system can support more multicast services and improve the expansion performance of the system.

The embodiment of the present invention also provides a message transmission system. As shown in FIG. 9, the message transmission system 900 includes: a centralized network node 901 and more than two edge routing bridge devices 902;

Wherein, the centralized network node 901 is in the same network as the above-mentioned two or more edge routing bridge devices 902;

The centralized network node 901 may be any network node as shown in FIGS. 6 to 8 , and its specific description may refer to the description in FIGS. 6 to 8 , which will not be repeated here.

It should be noted that, for the sake of simplicity of description, the aforementioned method embodiments are expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence. Because of the present invention, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification belong to preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and other media that can store program codes.

A message transmission method, network node, and message transmission system provided by the present invention have been introduced in detail above. In the above-mentioned embodiments, the descriptions of each embodiment have their own emphasis, and there is no detailed description in a certain embodiment. For the part, reference may be made to the relevant descriptions of other embodiments. For those of ordinary skill in the art, based on the ideas of the embodiments of the present invention, there will be changes in the specific implementation and application scope. In summary, the content of this specification should not be construed as limiting the present invention.

Claims (15)

1. A message transmission method, characterized in that, comprising: The centralized network node receives the message sent by the source edge routing bridge device through the multi-link transparent interconnection TRILL unicast encapsulation mode, wherein the message is a broadcast message or an unknown unicast message or a multicast message; The centralized network node determines a target edge routing bridge device set according to the multicast forwarding table, wherein the target routing bridge device set does not include the source edge routing bridge device, and the multicast forwarding table includes Identification of each edge routing bridge device in the network; The centralized network node sends the packets one by one to the routing bridge devices included in the target edge routing bridge device set through TRILL unicast encapsulation. 2. The method of claim 1, wherein, Before determining the target edge routing bridge device set according to the multicast forwarding table, the method also includes: receiving a link status message of each edge routing bridge device in the network where the centralized network node is located; The multicast forwarding table is generated according to the received link state message of each edge routing bridge device in the network where the centralized network node is located. 3. The method of claim 2, wherein, The link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device, and the multicast forwarding table specifically includes: the identifier and the ID of each edge routing bridge device in the network where the centralized network node is located. Correspondence between virtual local area network identifiers; The determining the target edge routing bridge device set according to the multicast forwarding table includes: Acquire the virtual local area network identifier in the message, and determine the edge routing bridge device that accesses the virtual local area network where the message is located according to the virtual local area network identifier and the multicast forwarding table; Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the packet is located. 4. The method of claim 2, wherein, The link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device and the identifier of the layer-2 multicast domain joined by the edge routing bridge device, then the multicast forwarding table specifically includes: the centralized The corresponding relationship between the identifier of each edge routing bridge device in the network where the network node is located, the virtual local area network identifier, and the identifier of the layer 2 multicast domain that the edge routing bridge device joins; The determining the set of target edge routing bridge devices according to the multicast forwarding table includes: Obtain the virtual local area network identifier and the identifier of the layer-2 multicast domain in the message, and use the combination of the identifier of the virtual local area network and the identifier of the layer-2 multicast domain as an index to determine the access location according to the multicast forwarding table the virtual local area network where the message is located, and join the edge routing bridge device of the Layer 2 multicast domain to which the message is accessed; Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the message resides and join the Layer 2 multicast domain that the message accesses. 5. The method according to any one of claims 1 to 4, wherein the method further comprises: The centralized network node floods the extended link state message in the network, the extended link state message carries the attribute and the superclass name of the centralized network node, and the attribute indicates that the centralized network node supports centralized Replication function, the super-level name is an alias used by the centralized network node in a centralized replication scenario. 6. The method of claim 5, wherein, The centralized network node receives the message sent by the source edge routing bridge device through the multi-link transparent interconnection TRILL unicast encapsulation mode, and then includes: Judging whether the egress alias of the message sent by the TRILL unicast encapsulation method is the super-level name, if the egress alias is the super-level name, perform the step of determining the set of target edge routing bridge devices according to the multicast forwarding table . 7. The method according to any one of claims 1 to 4, characterized in that, If the target edge routing bridge device set contains a virtual routing bridge device virtualized by multiple routing bridge devices with dual-homed access or multi-homed access, then the message is sent to the target edge one by one The routing bridge devices included in the routing bridge device collection include: Send the message to one of the routing bridge devices of the dual-homing access or multi-homing access; or, and sending the message to a virtual routing bridge device composed of multiple routing bridge devices with dual-homing access or multi-homing access. 8. A network node, characterized in that, comprising: The first receiving unit is used to receive the message sent by the source edge routing bridge device through the TRILL unicast encapsulation mode of transparent interconnection of multiple links, wherein the message is a broadcast message or an unknown unicast message or a multicast message; A determining unit, configured to determine a target edge routing and bridge device set according to a multicast forwarding table, wherein the target routing and bridge device set does not include the source edge routing and bridge device, and the multicast forwarding table includes the network where the network node is located The identity of each edge routing bridge device in ; The first sending unit is configured to send the packets received by the receiving unit one by one to the routing bridge devices included in the target edge routing bridge device set determined by the determining unit through TRILL unicast encapsulation. 9. The network node according to claim 8, characterized in that, The network node also includes: The second receiving unit is configured to receive a link state message of each edge routing bridge device in the network where the network node is located; A generating unit, configured to generate the multicast forwarding table according to the link state message of each edge routing bridge device in the network where the network node is located received by the second receiving unit. 10. The network node according to claim 9, characterized in that, The link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device, and the multicast forwarding table specifically includes: the identifier and virtual local area network identifier of each edge routing bridge device in the network where the network node is located. Correspondence between LAN identifiers; The determining unit is specifically configured to: obtain the virtual local area network identifier in the message, and determine the edge routing bridge device that accesses the virtual local area network where the message is located according to the virtual local area network identifier and the multicast forwarding table; Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the packet is located. 11. The network node according to claim 9, characterized in that, The link state message includes: the virtual local area network identifier of the virtual local area network accessed by the edge routing bridge device and the identifier of the layer 2 multicast domain accessed by the edge routing bridge device, then the multicast forwarding table specifically includes: the The corresponding relationship between the identifier of each edge routing bridge device in the network where the network node is located, the virtual local area network identifier, and the identifier of the layer 2 multicast domain that the edge routing bridge device joins; The determining unit is specifically configured to: obtain the virtual local area network identifier and the identifier of the layer-2 multicast domain in the message, and use the combination of the identifier of the virtual local area network and the identifier of the layer-2 multicast domain as an index, according to the The multicast forwarding table determines the access to the virtual local area network where the message is located, and joins the edge routing bridge device of the Layer 2 multicast domain that the message accesses; Wherein, the set of target edge routing bridge devices is composed of the determined edge routing bridge devices that access the virtual local area network where the message resides and join the Layer 2 multicast domain that the message accesses. 12. The network node according to any one of claims 8 to 11, characterized in that, The network node also includes: The second sending unit is configured to flood an extended link state message in the network, the extended link state message carries the attribute and the superclass name of the network node, and the attribute indicates that the network node supports centralized Replication function, the super-level name is an alias used by the network node in a centralized replication scenario. 13. The network node according to claim 12, characterized in that, The network node also includes: A judging unit, configured to judge the egress alias of the message sent by the TRILL unicast encapsulation mode when the first receiving unit receives the message sent by the source edge routing bridge device through the multi-link transparent interconnection TRILL unicast encapsulation mode Whether it is the name of the super class; The determining unit is triggered when the judging result of the judging unit is yes. 14. The network node according to any one of claims 8 to 10, characterized in that, The first sending unit is further configured to: when the set of target edge routing bridge devices includes a virtual routing bridge device virtualized by multiple routing bridge devices with dual-homing access or multi-homing access, send the One of the multiple routing bridge devices with homed access or multi-homed access sends the message, or sends the message to the multiple routing bridge devices with dual-homed or multi-homed access. The virtual routing bridge device sends the packet. 15. A message transmission system, comprising: Concentrate network nodes and more than two edge routing bridge devices; Wherein, the centralized network node is in the same network as the two or more edge routing bridge devices; The centralized network node is the network node according to any one of claims 8 to 14.